Pubertal Staging

For both sexes, the genital and pubic hair changes that unfold at puberty are classified into five stages: stage 1 is prepubertal and stage 5 is adult ( Fig. 17.1 ; Table 17.1 ). These physical changes may either be the result of gonadarche (as in the case of breast or testicular enlargement) or adrenarche (as in the case of pubic hair development). Although the physical sequelae of gonadarche and adrenarche generally occur concomitantly, discordance of the two processes may also occur in normal development.

Pubertal rating according to Tanner stages.

(A) Breast development in girls is rated from 1 (prepubertal) to 5 (adult). Stage 2 breast development (appearance of breast bud) marks the onset of gonadarche. For girls, pubic hair stages are rated from 1 (prepubertal) to 5 (adult). Stage 2 marks the onset of adrenarche. (B) Genital development in boys is rated from 1 (prepubertal) to 5 (adult). Stage 2 genital development, characterized by enlargement of the testes and scrotum accompanied by a change in the texture of the scrotal skin, marks the onset of gonadarche. Pubic hair development in boys is rated from stage 1 (prepubertal) to stage 5 (adult). Stage 2 represents the onset of pubarche, which can reflect adrenarche or gonadarche. Although pubic hair and genital or breast development are represented as synchronous in the figure, they do not necessarily take place simultaneously and should be scored separately. In normal boys, stage 2 pubic hair generally develops 1 to 1.5 years after stage 2 genital development.

(From Carel JC, Léger J: Clinical practice. Precocious puberty. N Engl J Med 358:2366, 2008. Reproduced with permission of the New England Journal of Medicine.)

During puberty, increased ovarian estrogen secretion promotes breast development in girls. The development of breast buds with increased areolar diameter is considered to be stage 2; greater enlargement of the breasts occurs in stage 3, accompanied by increased pigmentation of the areolae and nipples. During stage 4, the areolae are mounded above the breast tissue. Recession of the areola to the general breast contour represents breast stage 5. Palpation of the breast is necessary to better differentiate between breast tissue and lipomastia. Additional effects of estrogen at this stage of development include cornification of the vaginal mucosa, uterine growth, and morphogenesis of an adult female body habitus.

Menarche follows an anovulatory cycle and generally occurs 2 to 3 years after the onset of breast development. Menstrual cycles during the first year after menarche are typically irregular and anovulatory, with most ranging in duration from 21 to 45 days. By 3 years postmenarche, over 90% of adolescent females have 10 or more menstrual cycles/year with an average menstrual interval of 36.5 days. Nevertheless, cycles can remain irregular until the fifth year postmenarche.

Although primordial and preantral follicles predominate during the prepubertal years, small antral follicles can develop during this phase of maturation. These small follicles are gonadotropin independent. Ovarian volume increases with the onset of puberty, achieves maximum volume soon after (between menarche and age 16 years), and remains stable or decreases slightly thereafter. Polycystic ovary morphology (PCOM) is frequently detected in healthy adolescent girls; this morphology is not associated with decreased ovulatory rate, hyperandrogenism, or metabolic abnormalities. During the early postmenarcheal period, ovarian morphology on transabdominal ultrasound shows multicystic ovaries and increased ovarian volume that differ from ovarian morphology observed in older women.

In girls, increased adrenal androgen secretion is considered to be responsible for the development of darker hairs along the labia, which is classified as pubic hair stage 2. The hair becomes darker and coarser during pubic hair stage 3, spreading over the pubic symphysis with gradual progression to a full female escutcheon. Apocrine odor may precede or accompany the development of pubic hair. Associated findings include axillary hair, acne, and oiliness of skin and hair.

For boys, an increase in testicular volume and enlargement of the scrotum is considered genital stage 2. At stage 2, the testes are approximately 4 to 8 mL in volume with the longest axis being approximately 2.5 cm. The volume of the mature human testis is approximately 20 to 30 mL and represents increased growth of the seminiferous tubule due to Sertoli cell proliferation and differentiation as well as initiation of spermatogenesis. At genital stage 3, further growth of the testes has occurred, and the length and diameter of the penis has increased. At genital stage 4, penile size has increased and the scrotal skin has become darkened. Palpation and use of orchidometer is preferable to inspection. Male pubic hair stage 2 consists of downy hair at the base of the penis. For pubic hair stage 3, the hair is longer, darker, and extends over the junction of the pubic bones. For pubic hair stage 4, the extent of hair has increased, but has not yet achieved the adult male escutcheon. Other secondary sexual characteristics in boys include axillary hair, increased size of the larynx, deepening of the voice, increased bone mass, and increased muscle strength. Approximately 3 years after the appearance of pubic hair, terminal hair appears in androgen-dependent regions on the face and trunk where it may develop for several years thereafter. There is considerable variation in the distribution and density of beard, chest, abdominal, and back hair, presumably reflecting genetic differences. The appearance of spermatozoa in early morning urine specimens (spermaturia) occurs during genital stage 3. Gynecomastia is observed in 50% of boys. Typically this is most prominent in midpuberty when the ratio of circulating concentrations of estradiol to testosterone is relatively high. In most instances, gynecomastia resolves spontaneously by 16 years of age.

The pubertal growth spurt in girls occurs concurrently with the onset of breast development. Usually only 4 to 6 cm of growth occur after menarche. The pubertal growth spurt in boys, with an average height velocity of 9.5 cm per year, occurs around genital stages 3 and 4. In general, the age at peak height velocity shows an inverse relationship with the magnitude of the growth spurt. Linear growth is approximately 99% complete for girls at a bone age of 15 years and for boys at a bone age of 17 years.

Breast development in girls and testicular enlargement in boys generally precede pubic hair development. Yet, the tempo for pubic hair development is faster such that synchrony between genital and pubic hair development occurs during the later stages of puberty. Schemata for the temporal development of the secondary sexual characteristic and their relationship to growth velocity is shown for girls and boys in Figs. 17.2 and 17.3 , respectively.

Mean ages (dots) and ranges (horizontal lines) of pubertal onset and development in girls.

Dashed lines show the earlier limit for African-American girls. Max , Maximum.

(From Lee PA: Puberty and its disorders. In Lifshitz F, editor: Pediatric Endocrinology , ed 4, New York, 2003, Marcel Dekker, p 212. Reproduced with permission of Dr. Peter Lee and Marcel Dekker.)

Mean ages (dots) and ranges (horizontal lines) of pubertal onset and development in boys.

Dashed lines show the earlier limit for black boys.

(From Lee PA: Puberty and its disorders. In Lifshitz F, editor: Pediatric Endocrinology , ed 4, New York, 2003, Marcel Dekker, p 213. Reproduced with permission of Dr. Peter Lee and Marcel Dekker.)

Secular Trends and Racial and Ethnic Differences in the Onset and Tempo of Puberty

Reports of secular changes in the onset of puberty have focused on girls and have typically used age at menarche as the biomarker for puberty.

Gluckman and Hanson have suggested that menarche occurred between 7 and 13 years in Paleolithic and Neolithic times. Based on analysis of 994 medieval adolescent skeletons (10 to 25 years) for pubertal stage, it appears that adolescents began puberty around 10 to 12 years with menarche occurring closer to 15 years in rural areas and 17 years in London. Boys experienced protracted pubertal stages during the 10th to 17th centuries. Potential factors that contributed to later puberty during medieval times included poor diet, increased infections, and greater physical exertion.

Available historical data indicate that this was followed by a decline in the age of menarche in Europe and North America from the early 19th century (16 to 17 years of age) to the latter half of the 20th century (13 years of age). This trend has been attributed to the improving socioeconomic conditions during this epoch. Although recent data from North America, several European countries, and other regions of the industrialized world suggest that the trend to earlier menarche has been reduced or halted, breast and pubic hair development are apparently occurring earlier than 50 years ago in both North America and Europe.

The biology underlying this continued positive secular trend in sexual development in girls, which in some populations is loosely associated with a similar trend in growth, is unclear, and may or may not involve an earlier onset of gonadarche or adrenarche. This earlier onset of breast development is not associated with increased gonadotropin or estradiol concentrations, suggesting that this represents a gonadotropin-independent event. Earlier breast development assessed by palpation was reported in NHW girls, which was likely related to the increased BMI of this group. Analogous studies of boys are limited, but no striking sex differences in secular trends in puberty and growth are apparent. Using both genital staging and the orchidometer, the Copenhagen Puberty Study reported pubertal onset occurring 3 months earlier in Danish boys and reported that obesity advanced the onset of testicular enlargement. However, other studies suggest that obesity delays onset of puberty in boys.

The age at onset of puberty varies between ethnic groups. Among American girls, mean ages for breast development, pubic hair development, and menarche were 9.5, 9.5, and 12.1, respectively, for NHB girls; 9.8, 10.3, and 12.2, respectively, for MXAM girls; and 10.3, 10.5, and 12.7 years, respectively, for NHW girls. Data obtained through the cross-sectional Third National Health and Nutrition Examination Survey (NHANES III) between 1988 and 1994 showed that NHB girls enter puberty first, followed by MXAM and NHW girls. Based on the NHANES III study, luteinizing hormone (LH) and inhibin B concentrations associated with onset of breast development were evaluated. As would be anticipated, LH and inhibin B concentrations increased with pubertal progression. Cut-points for Tanner 2 breast development were LH 1.04 mIU/mL (95% confidence interval [CI]: 0.71 to 1.37) and inhibin B 17.89 pg/mL (95% CI: 11.63 to 24.15). The respective median ages at hormonal onset based on LH concentrations were 10.7, 10.6, and 10.1 years for NHW, MXAM, and NHB girls, respectively. Girls with low birthweight and greater postnatal weight gain had relatively earlier onset of puberty based on LH concentrations and a similar pattern regarding pubertal onset was noted based solely on postnatal weight gain.

For boys in NHANES III, median estimated ages for genital and pubic hair stage 2 were 9.3 and 11.1, respectively, for NHB boys; 10.4 and 12.3, respectively, for MXAM boys; and 10.1 and 12.0, respectively, for NHW youths. For genital and pubic hair stage 5, median ages were 14.9 and 15.2, respectively, for NHB boys; 15.8 and 15.7, respectively, for MXAM boys; and 16.0 and 15.6, respectively, for NHW boys. Using the NHANES III data, LH, testosterone, and inhibin B concentrations increased as puberty progressed. Likely reflecting individual and diurnal variation, no single or combination hormone cut-point was found to be predictive of physical pubertal status, either genital or pubic hair stage 2, in this population.

The age at puberty reflects interactions between genetic, prenatal, and environmental factors. Twin studies indicate that heredity is responsible for approximately 50% of the variation in age at menarche. Indeed, pubertal timing of both parents has been demonstrated to influence the timing of pubertal onset of both boys and girls independent of sex. Investigation of genetic variants associated with onset of puberty identified a specific variant, −29G>A, in the promoter region of the follicle-stimulating hormone (FSH) receptor (FSHR) gene; breast development occurred 7.4 months later among homozygous carriers of the −29G>A variant compared to the −29GG+GA carriers.

Steroidogenesis

The biosynthetic pathways for gonadal and adrenal steroids are considered together because of their similarities and the importance in understanding the physiology and pathophysiology of puberty ( Fig. 17.4 ; see Chapter 4 ). Synthesis of steroids from cholesterol requires the expression of specific enzymes, receptors, co-factors, and other proteins in the adrenal cortex and the gonads. Steroidogenesis is regulated by specific trophic hormones, ACTH, LH, and FSH.

Steroidogenesis.

This figure shows pathways of adrenal, ovarian, and testicular steroidogenesis. Dark solid lines indicate predominant pathways for adrenal steroidogenesis. Dashed lines indicate predominant pathways for gonadal steroidogenesis. Genes responsible for specific enzymatic steps are indicated. Enzymatic steps for which P450 oxidoreductase serves as a co-factor (encoded by POR) are indicated. Arrow indicates “backdoor pathway” of steroidogenesis. DHEA , Dehydroepiandrosterone; DHEAS , dehydroepiandrosterone sulfate; DHT , dihydrotestosterone; DOC , deoxycorticosterone; HSD , hydroxysteroid dehydrogenase.

(Modified from Witchel SF, Lee PA: Ambiguous genitalia. In Sperling MA, editor: Pediatric Endocrinology , ed 2, Philadelphia, 2002, WB Saunders, p 119.)

The adrenal cortex consists of three zones, the zona glomerulosa (ZG), zona fasciculata (ZF), and zona reticularis (ZR). The ZG synthesizes aldosterone, a mineralocorticoid, and is primarily regulated by potassium concentrations and renin-angiotensin. The ZF synthesizes cortisol. Steroidogenesis in the ZF is primarily governed by ACTH. The ZR synthesizes C-19 steroids, such as DHEA, DHEAS, androstenedione, androstenediol, and 11β-hydroxyandrostenedione.

ACTH is a peptide derived following proteolytic cleavages of proopiomelanocortin (POMC). Its actions are mediated by the ACTH receptor, a seven-transmembrane G protein-coupled receptor encoded by MC2R. This pathway utilizes cyclic adenosine monophosphate (cAMP)-dependent protein kinase A. The acute effects of ACTH include uptake of plasma low-density lipoproteins, stimulation of cholesterol esterase activity, enhanced synthesis and phosphorylation of steroidogenic acute regulatory protein (StAR), cholesterol transfer across the inner mitochondrial membrane, and increased cortisol secretion. The chronic effects of ACTH involve stimulation of transcription and translation of steroidogenic enzyme genes.

In the gonads, LH and FSH modulate steroid biosynthesis. LH promotes ovarian theca cell and testicular Leydig cell steroidogenesis; its actions are mediated by its cognate receptor, LHCGR. FSH acting through the FSHR stimulates aromatase expression to promote estrogen biosynthesis in the ovary and Sertoli cell growth in the testis. The LH and FSH receptors are both G protein-coupled receptors and contain leucine-rich repeats in their large ectodomains.

Most enzymes involved in steroidogenesis are cytochrome P450s (CYPs) or hydroxysteroid dehydrogenases (HSDs). The rate-limiting step of steroidogenesis is transport of cholesterol into mitochondria mediated by StAR. Within the mitochondria, cholesterol desmolase (also known as side-chain cleavage or P450scc) converts cholesterol into pregnenolone. One enzyme, 17α-hydroxylase/17,20-lyase (P450c17), encoded by the CYP17A1 gene, is the qualitative regulator of adrenal and gonadal steroidogenesis. This enzyme mediates 17α-hydroxylation to convert pregnenolone into 17α-hydroxypregnenolone. In the ZR, ovarian theca, and Leydig cells, this same enzyme catalyzes scission of the C17–20 bond to produce DHEA. Although this one protein is capable of two distinct enzymatic reactions, these enzyme activities are differentially regulated. Factors known to modulate 17,20-lyase activity include: (1) the amount of P450 oxidoreductase (POR); (2) the expression of cytochrome b ₅ (CYB5A); (3) the phosphorylation of serine/threonine residues on P450c17; and (4) the phosphorylation of noncanonical P450c17 residues. POR is a protein that transfers electrons from nicotinamide adenine dinucleotide phosphate to microsomal cytochrome P450 enzymes, such as P450c17, P450c21, and aromatase (P450aro). CYB5A modulates adrenal androgen secretion by increasing the 17,20-lyase activity of (P450c17).

The Δ -steroids are converted to the Δ -steroids by 3β-hydroxysteroid dehydrogenase type 2 (HSD3B2), the adrenal and gonadal specific isoform. This enzyme converts pregnenolone to progesterone in the ZG, 17-hydroxypregnenolone to 17-hydroxyprogesterone in the ZF, and DHEA to androstenedione in the ZR. In the ZF, 17-hydroxyprogesterone (17-OHP) is converted to 11-deoxycortisol by 21-hydroxylase (P450c21) and, subsequently, to cortisol by 11β-hydroxylase (P450c11β).

The adrenals, ovaries, and testes synthesize sex steroids. The ZR of the adrenal cortex produces DHEA, DHEAS, androstenedione, androstenediol, and 11β-hydroxyandrostenedione. DHEA sulfotransferase (SULT2A1) converts DHEA to DHEAS; this enzyme is also expressed in the liver. Sulfation of steroids by SULT2A1 requires a sulfate donor, 3′-phosphoadenosine 5′-phosphosulfate (PAPS), and the enzyme PAPS synthase. In the ovary, androstenedione is synthesized in the theca cell and diffuses into the granulosa cell, where it is aromatized by aromatase (P450aro) to estrone and converted to estradiol by 17β-hydroxysteroid dehydrogenase type 1 (HSD17B1). In Leydig cells, androstenedione is converted to testosterone by HSD17B3. In many androgen target cells, such as those in the external genitalia and prostate, testosterone is converted to DHT by 5α-reductase type 2. In other androgen-sensitive tissues, such as bone and adipose, testosterone is converted to estradiol by aromatase. Whereas DHT is the most potent ligand for androgen receptor (AR), some adrenal C-19 steroids can undergo peripheral conversion to hormones capable of binding to the AR, which is encoded by the AR (NR3C4) gene.

The 17β-hydroxysteroid dehydrogenase enzymes comprise a large family of enzymes involved in steroid biosynthesis and metabolism. The differences in tissue distribution, substrate preferences, subcellular localization, and mechanisms of regulation influence the cellular steroid micro-environment. The type 1 isozyme, 17βHSD1, is expressed in ovaries, placenta, endometrium, and liver, where it favors conversion of estrone to estradiol. The type 3 isozyme, 17βHSD3, is expressed in the testis, where it preferentially converts androstenedione to testosterone. The type 5 enzyme, 17βHSD5, is an aldo-keto-reductase (AKR) enzyme (AKR1C3) that is expressed in steroidogenic and peripheral tissues; it can convert androstendione to testosterone.

Through investigations of the tammar wallaby and patients with disordered steroidogenesis, the presence of another pathway leading to dihydrotestosterone (DHT) synthesis was elucidated. In this “alternative backdoor pathway,” 17-OHP undergoes 3α- and 5α-reduction followed by 17,20-lyase, 17β-hydroxysteroid dehydrogenase, and 3α-oxidation steps to generate DHT in the absence of the “classic” intermediates, DHEA, androstenedione, and testosterone. In humans, since 17-OHP is not a favorable substrate for the 17,20-lyase reaction, this pathway acquires functional importance in disorders of steroidogenesis associated with increased 17-OHP concentrations, such as congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency and oxidoreductase deficiency.

Once secreted, sex steroids circulate bound to sex hormone-binding globulin (SHBG) and to albumin. The unbound or free hormone is the bioavailable form that diffuses passively into target cells and interacts with nuclear steroid receptors. Steroid hormone receptors are ligand dependent transcription factors, comprised of three functional domains: the N-terminal domain serves to modulate function; the DNA-binding domain mediates binding of the receptor to DNA; and the ligand-binding domain binds to steroid. Steroid receptor activity is modulated by various tissue-specific cofactors; both coactivators and corepressors can influence receptor function.

Steroids also act through nongenomic mechanisms. For example, testosterone can activate phospholipase C, leading to calcium influx into Sertoli cells, and can activate the mitogen-activated protein kinase as well as other intracellular mediators. Sex steroids can be metabolized to inactive forms by a variety of enzymes. Glucuronidation decreases the biological activity of steroid hormones and increases solubility to facilitate renal excretion. This process, catalyzed by UDP-glucuronyltransferase (UGT) enzymes, involves the transfer of glucuronic acid from uridine diphosphoglucuronic acid to steroid hormones. In humans, the UGT2B isoforms show greater specificity for C19 androgens. A second mechanism is sulfoconjugation, in which DHEA sulfotransferase catalyzes conversion of DHEA to DHEAS, and estrogen sulfotransferase converts estrogens to estrone sulfate. The inactive sulfated steroids can be hydrolyzed to active forms by steroid sulfatase.

Activation of Gonadarche

Gonadarche reflects increased activity of the hypothalamic-pituitary-gonadal (HPG) axis (see Chapter 1 ). The increased pituitary secretion of LH and FSH stimulates gonadal steroidogenesis, development of the physical manifestions of puberty, completion of gametogenesis, and maintenance of fertility. LH and FSH are heterodimeric proteins comprising a common α-subunit and a unique β-subunit; both are glycosylated peptides. Glycosylation appears to modulate hormone stability, protein folding, cellular trafficking, circulating serum half-life, and receptor signaling. The temporal increments in circulating LH and FSH concentrations at the time of puberty, and their relationships to those of the gonadal steroids, testosterone and estradiol, respectively, at this stage of development, have been well documented in both boys and girls. The actions of LH and FSH are mediated by their cognate seven-transmembrane domain G protein-coupled receptors: the LH receptor (LHCGR) and FSHR, respectively.

In the female, LH stimulates androgen production by the thecal cells of the ovarian follicles and progesterone secretion from luteinized granulosa cells of the corpus luteum. FSH is critical for the process of follicular recruitment and selection. In granulosa cells of the developing follicle, FSH induces expression of aromatase, which is responsible for aromatization of the theca-cell-derived androgens into estrogens. FSH also induces LHCGR expression in granulosa cells of the dominant follicle, which selectively amplifies the effect of declining FSH concentrations on the dominant follicle. In the male, LH regulates the secretion of testosterone from Leydig cells.

FSH, together with testosterone, is responsible for initiating and maintaining spermatogenesis. The action of FSH in this regard is indirect and exerted on the somatic Sertoli cell of the seminiferous tubule. While the actions of testosterone are also indirect, several somatic cell types in the testis (Sertoli, Leydig, and peritubular) express AR and are considered to be involved in the control of spermatogenesis.

The pubertal drive to the pituitary-gonadal axis is generated by a diffusely distributed network of hypothalamic neurons expressing GnRH-1, known as the “hypothalamic GnRH pulse generator.” As the name implies, the hypothalamic GnRH pulse generator produces intermittent discharges of GnRH into the hypophysial portal circulation, which is obligatory for gonadotropin synthesis and secretion by the pituitary gonadotrophs. LH and FSH secretion is stimulated by GnRH acting through its receptor, GnRH-R1, located on gonadotropin-secreting cells (gonadotrophs) in the pituitary gland. A unique feature of the human GnRHR, a 7-transmembrane domain G protein-coupled receptor, is its lack of a C-terminal cytoplasmic domain.

A second GnRH gene (GnRH-II) is also expressed by neurons in the primate brain and a GnRH-II receptor has been cloned from the pituitary, but the gene encoding this receptor does not generate a functional protein in humans. The significance, if any, of this second GnRH system in the control of the human pituitary-gonadal axis has not been delineated. Similarly, a hypothalamic RF amide-related peptide (RFRP3) that is inhibitory to gonadotropin secretion in several species and considered to be the homolog of gonadotropin-inhibitory hormone (GnIH) in birds has been identified in mammalian species, GnIH neurons are mainly concentrated in the dorsomedial hypothalamus, form appositions with some GnRH neurons, and signal through its cognate receptor, GPR147. Nevertheless, the potential roles, if any, of GnIH and GPR147 on gonadotropin secretion in man and nonhuman primates remain to be clarified.

Other factors have been studied during puberty. These factors include inhibins, activins, anti-müllerian hormone, insulin-like factor-3 (INSL3), and osteocalcin. Activins and inhibins are members of the transforming growth factor-β (TGF-β) superfamily composed of a common α-subunit and two β-subunits (β _A and β _B ). The activins are dimers consisting of only the β-subunits; activin A is a dimer of β _A subunits and activin B is a dimer of β _B subunits. The activins are synthesized in the gonadotropes and influence FSH secretion. Mature inhibins are dimers composed of a common α-subunit covalently linked with one of two β-subunits (β _A and β _B ). The α/β _A and α/β _B dimers are known as inhibin A and inhibin B, respectively. The gonadal inhibins, like the gonadal steroids, play both an endocrine role in the regulation of gonadotropin secretion and a paracrine role within the gonads. Inhibin B, which is synthesized in part by the Sertoli cell, is the principal inhibin secreted by the testis. Inhibin B concentrations are low in prepubertal boys and increase with the onset of puberty. The pubertal increase in inhibin B may be attributed to Sertoli cell proliferation and to the initiation of spermatogenesis, both of which reflect the increased gonadotropin drive to the testis at the time of gonadarche.

In girls, circulating levels of inhibin A and B are low or undetectable prior to puberty. Inhibin B begins to rise with the onset of puberty, as does inhibin A in breast stages 3 and 4. Adult levels are attained at approximately 14 to 15 years of age. During the menstrual cycle, inhibin A levels are elevated in the luteal phase, while inhibin B predominates in the circulation of the follicular phase. However, the role of ovarian inhibins in regulating gonadotropin secretion in pubertal and premenopausal women remains to be fully elucidated.

Osteocalcin is secreted by osteoblasts. Most available data regarding the potential actions of osteocalcin are derived from mouse studies. Studies of mice have shown that osteocalcin facilitates testicular testosterone secretion and pancreatic β-cell proliferation. The testicular effects of osteocalcin appear to be independent of the HPG axis as the osteocalcin receptor is not expressed in the hypothalamus or pituitary of mice. Posttranslationally, osteocalcin is carboxylated on three glutamic acid residues. The carboxyated form of the molecule is considered to be biologically inactive whereas the undercarboxylated form is biologically active. Osteocalcin signals through its cognate receptor, GPRC6A, which is expressed by Leydig cells. Interestingly, increased osteocalcin concentrations are associated with the pubertal rise in testosterone concentrations in boys. Osteocalcin shows sexual dimorphism because it modulates Leydig cell testosterone production, but not ovarian estrogen production. The discovery of a missense mutation in the GPRC6A gene in two men with primary testicular failure, oligospermia, and glucose intolerance encourages speculation that osteocalcin influences testicular function in humans.

Anti-müllerian hormone (AMH), also known as müllerian-inhibiting hormone (MIH), is another member of the TGF-β superfamily. AMH signals by binding to a specific type-II receptor (AMHR2); this receptor heterodimerizes with one of several type-I receptors (ALK2, ALK3, and ALK6), leading to recruitment of Smad proteins that are translocated to the nucleus to regulate target gene expression. AMH is secreted by the Sertoli cells of the developing testis and stimulates regression of the müllerian ducts during male fetal development. Postnatally, AMH is secreted by the Sertoli cells of the testis. In boys, AMH concentrations decline at puberty. Although FSH stimulates AMH production, testosterone inhibits Sertoli cell AMH secretion. The decline in AMH concentrations appears to be closely coupled to rising inhibin B concentrations at the onset of puberty in boys—the latter presumably reflecting androgen induced differentiation of the Sertoli cell. Both AMH and inhibin B concentrations are low in boys with bilateral anorchia or complete hypogonadotropic hypogonadism (HH).

Immunohistochemical studies of human ovaries showed no AMH staining in primordial follicles, highest expression in growing preantral and small antral follicles, and disappearance in larger follicles (>8 mm). In ovaries, AMH is primarily secreted by the granulosa cells of the preantral and antral follicles. During infancy, AMH concentrations increase to achieve a plateau during adolescence until age 25 years. Subsequently, AMH concentrations decline and correlate inversely with age. AMH plays a role as a gatekeeper of follicular development.

Data obtained from studies in rodents have demonstrated presence of the AMH receptor, AMHRII, in gonadotropes. One report using rats showed that AMH stimulated FSH secretion in immature female rats. Subsets of GnRH neurons express the AMH receptor. In vitro studies have demonstrated that AMH increases GnRH-dependent LH pulsatility and secretion. Thus AMH may influence gonadotropin secretion.

INSL3, a peptide hormone secreted by Leydig cells, plays a major role in directing the transabdominal phase of testicular descent during gestation. INSL3 signaling is mediated by the G protein-coupled relaxin family peptide receptor 2 (RXFP2). After birth, fetal Leydig cells involute followed by quiescence until puberty. Adult-type Leydig cells derive from a different population of precursors and are dependent on LH stimulation. INSL3 concentrations are low during infancy, rise during puberty, and reflect Leydig cell function during adulthood. Longitudinal data demonstrated that INSL3 concentrations rise with onset of puberty and increasing testicular volume.

INSL3 is also secreted by ovarian theca cells, predominantly by theca cells surrounding medium and large growing follicles. INSL3 and its receptor may play a role in autoregulatory feedback to maintain theca cell androgen production. Despite much variation, INSL3 concentrations tend to rise during late puberty in girls; the variation likely reflects that INSL3 is secreted largely by growing follicles and is a marker of theca cell activity.

In the pubertal and postpubertal individual, the ovaries and testes are governed by feedback control systems. GnRH and the gonadotropins comprise the feed-forward components from hypothalamus to pituitary and from pituitary to gonad, respectively. Steroid and protein hormones from the gonads, in turn, provide the feedback signals that regulate the secretion of LH and FSH. The feedback actions of these gonadal hormones, which involve both negative and stimulatory (positive) actions, may be exerted directly at the level of the pituitary gonadotrophs to modulate expression of the genes encoding LHβ and FSHβ ( LHB and FSHB , respectively). Feedback may also be exerted indirectly at the level of the hypothalamus to regulate the release of GnRH.

In the male, a negative feedback action of testosterone and inhibin B are the major regulators of LH and FSH secretion, respectively. The action of testosterone is predominantly exerted at the hypothalamic level, while that of inhibin appears to occur directly at the pituitary. The role of aromatization of testosterone to estradiol in mediating the negative feedback action of this androgen on LH secretion continues to be an area of active investigation. The feedback control of LH and FSH throughout the menstrual cycle is complicated and involves both negative and positive feedback actions of ovarian steroids at both the hypothalamic and pituitary levels (see also Chapter 8 ). The maintenance of normal ratios of circulating LH and FSH concentrations is important for gonadal function, particularly for folliculogenesis and ovulation.

In conditions under which pulsatile GnRH release is compromised, such as occurs in anorexia nervosa and during periods of strenuous physical training especially in young women, gonadotropin secretion is attenuated and pubertal development is arrested. Thus the pituitary-gonadal axis in both males and females may be viewed as being a slave to the hypothalamic GnRH pulse generator, and this analogy should be held in mind when considering the mechanisms triggering the onset of gonadarche.

Hypothalamic Gonadotropin-Releasing Hormone Pulse Generator

The human hypothalamus contains diffusely distributed GnRH neurons, many of which send their projections to the median eminence, where they synchronously and intermittently discharge their peptide into the primary plexus of the hypophysial portal circulation, thereby providing the pituitary gonadotrophs with the pulsatile stimulation that is essential for maintaining gonadotropin secretion. While the neurobiological mechanisms that underlie GnRH pulse generation remain controversial, compelling evidence indicates the fundamental role of KNDy neurons in the arcuate nucleus (a.k.a. infundibular nucleus). These hypothalamic neurons, named because they coexpress kisspeptin, NKB, and dynorphin, project their axons to the median eminence where they mingle intimately with GnRH fibers en route to the portal vessels. The GnRH fibers that target the median eminence display unique features in that these fibers exhibit characteristics of both axons and dendrites and have been termed “dendrons.” At the median eminences, these fibers are intertwined, encased by tanycytes (specialized ependymal cells of the third ventricle), project to multiple blood vessels, and receive numerous synaptic inputs.

Kisspeptin is an extremely potent GnRH secretagogue and signals through its cognate receptor, KISS1R, which is expressed in GnRH neurons. Kisspeptin fibers project onto GnRH cell bodies and GnRH fibers. Using immunohistochemistry, kisspeptin was detected in the anterior and intermediate lobes of the pituitary in monkeys, but was not apparently colocalized with gonadotrophs, somatotrophs, or lactotrophs. Conflicting data exist regarding the functional relevance of direct kisspeptin action at the level of the pituitary gland. Nevertheless, available evidence suggests that bidirectional central and peripheral signaling modulate reproductive function. Use of Gpr54 KO mice with selective re-introduction of Gpr54 into GnRH cells confirmed that direct effects of kisspeptin on GnRH cells are sufficient to attain fertility, but insufficient to preserve normal functionality of the reproductive axis; kisspeptin does not signal directly at the level of the pituitary.

It has been proposed that GnRH pulse generation is achieved by reciprocal stimulatory (NKB) and inhibitory (dynorphin) connections within the arcuate nucleus, while the output of the pulse generator is relayed to GnRH fibers projecting to the median eminence by an intermittent kiss­peptin signal. Thus these neurons may represent the anatomic site of the GnRH pulse generator. Loss-of-function mutations in the genes encoding for kisspeptin (KISS1), the kisspeptin receptor (KISS1R), NKB (TAC3), or the NKB receptor (TACR3) in humans are associated with hypogonadotropic states that are typically manifest at puberty. The importance of these neurons was confirmed by the demonstration that selective ablation of KNDy neurons in the postpubertal rat results in a loss in hypothalamic drive to LH secretion.

The KNDy hypothesis of the GnRH pulse generator was investigated in adult male volunteers by administering kisspeptin-54, NKB, and an opioid receptor antagonist, naltrexone. LH pulsatility was used as a surrogate marker for GnRH pulsatility. Kisspeptin alone potently increased LH and LH pulsatility, which was consistent with previous observations in humans. NKB alone, on the other hand, did not affect gonadotropins. Coadministration of NKB and kisspeptin had significantly lower increases in gonadotropins compared with kisspeptin alone. The combination of naltrexone and kisspeptin significantly increased LH pulse amplitude. These results suggest that interactions between kisspeptin, NKB, and dynorphin influence LH pulsatility and gonadotropin release in humans. In women, NKB appears to influence GnRH/LH secretion in normal women through mechanisms predominantly proximal to kisspeptin in mediating estrogen positive and negative feedback on LH secretion. Overall, these studies have begun to elucidate functional aspects regarding the neurobiology of KNDy neurons. Clarifying the details of the potential interaction of kisspeptin, dynorphin, and NKB warrants further studies.

Curiously, a role for kisspeptin (beyond the GnRH pulse generator) has been implicated as integrating behavior and emotions with reproduction in humans in that kisspeptin administration enhanced limbic brain activity captured on functional MRI in response to sexual images and nonsexual bonding images in healthy volunteer men.

Coordination and interactions among various hypothalamic factors influence GnRH secretion by the GnRH neuron. Both excitatory and inhibitory transsynaptic neuronal inputs modulate the GnRH neuronal system. In addition to neuronal afferents, GnRH neurons maintain close physical contact with glial cells. Thus the secretory activity of GnRH neurons reflects the integrated response to hormones, neurotransmitters, neuromodulators, paracrine interactions, and environmental cues. Leptin, insulin, IGF-1, ghrelin, FGF21, orexigenic peptides, and anorexigenic peptides provide input signals that are deciphered and organized to govern pubertal maturation and ongoing reproductive function.

Fetal Development of the Gonadotropin-Releasing Hormone Pulse Generator

Using three-dimensional (3D) imaging and transparent human fetal brains, Casoni et al. have suggested that approximately 2000 GnRH neurons reside in the hypothalamus and approximately 8000 are widely distributed in other areas of the brain. Meticulously orchestrated development of the GnRH neurons and olfactory neurons in conjunction with precise spatio-temporal expression of multiple factors is essential for normal HPG axis function ( Fig. 17.5 ). Specific mutations in families with disorders of puberty and studies of transgenic mice have established some of the factors involved in GnRH neuron migration. Elements essential to GnRH neuron development, migration, and function include cytoskeletal proteins, adhesion molecules, neurotransmitters, growth factors, receptors, and transcription factors. Some factors function in multiple regions and may have diverging effects depending on the context of the molecular environment. Elucidation of the neurobiology and ontogeny of the GnRH neurons will improve understanding of the pathophysiology of HH and, perhaps, lead to novel therapies.

Genes implicated in gonadotropin-releasing hormone (GnRH) neuron development, GnRH neuron migration, and GnRH neuron activity and signaling.

Genes implicated in pituitary development and signaling are listed. Genes have been classified by presumed function in these processes.

The GnRH neurons are born in the olfactory placode and migrate during early fetal development from the nose through the forebrain to the hypothalamus. The fetal ontogeny of the GnRH neurons can be classified into several stages, each with distinct regulatory mechanisms: (1) differentiation of GnRH neurons; (2) axonophilic migration with axons of the vomeronasal nerve across the cribriform plate and into the forebrain; (3) localization in the hypothalamus and development of processes to the median eminence; and (4) attainment of final location and functionality.

Most vertebrates have two distinct olfactory systems: the main olfactory system responsible for recognition of volatile odorants and the vomeronasal system responsible for detection of pheromones. During the fifth week of gestation in humans, the olfactory placodes develop as thickenings of the ectoderm on the ventrolateral sides of the head. The GnRH progenitor stem cells include cells derived from this embryonic olfactory placode and cells derived from the neural crest.

Two proteins, chromodomain helicase DNA-binding protein 7 (CHD7) and SOX10, influence neural crest cell development and eventual migration. CHD7 is a large protein that participates in chromatin remodeling and transcription; it interacts with other proteins and may regulate genes involved in neural crest cell guidance. In mice heterozygous for Chd7 mutations, Fgfr1 expression in the olfactory placode, GnRH1 and Otx2 expression in the hypothalamus, and GnRHR expression in the pituitary were decreased, supporting a role for CHD7 upstream of Fgf8 and Fgfr1 in the development and maintenance of GnRH neurons.

Other factors involved in the differentiation of GnRH neurons include fibroblast growth factor-8 (FGF8), fibroblast growth factor receptor-1 (FGFR1), and heparan sulfate 6- O -sulfotransferase 1 (HS6ST1). FGF8 influences craniofacial development, neuroendocrine cell proliferation, cell fate specification, and cell survival. In the developing GnRH neuron, FGFR1 is the preferred receptor for FGF8. The FGFR1 is a tyrosine kinase receptor composed of three extracellular immunoglobulin domains, a transmembrane domain, and a cytoplasmic tyrosine kinase domain. Upon ligand binding, FGFR1 and its coreceptor dimerize, leading to autophosphorylation and protein kinase activity. In humans, FGF8 and FGFR1 loss-of-function mutations have been associated with holoprosencephaly, impaired forebrain cleavage, and midline facial anomalies, highlighting the important role of this signaling pathway.

The extracellular domain of FGFR1 interacts with heparan sulfate proteoaminoglycan, its coreceptor. Heparan sulfates are cell membrane and matrix-associated proteoglycans involved in neural development. These polysaccharides undergo nonrandom modifications of the sugar moieties to facilitate cell-to-cell communication. HS6ST1 introduces a sulfate at the 6- O position within heparan sulfate. This action appears to be necessary for FGFR1 function.

GnRH neuronal differentiation occurs between 39 and 44 days of gestation. Between weeks 5 and 6 of human gestation, the migratory mass of neural crest-derived migratory cells and olfactory neurons contains a small number of GnRH neurons. Subsequently, around the sixth week of gestation, the GnRH neurons begin their migration along vomeronasal nerves through the cribriform plate and eventually find their way to the hypothalamus. The GnRH neurons migrate into the brain along two distinct migratory pathways: a ventral pathway directed towards presumptive hypothalamic regions and a dorsal pathway directed toward pallial and subpallial telencephalic regions. These GnRH neurons, accompanied by olfactory ensheathing cells (OECs), travel along with the terminal, vomeronasal, and olfactory nerves into the brain. OECs are glial cells that guide GnRH and olfactory nerves to the forebrain. OECs express several factors important for GnRH migration, such as semaphorin 4D, signaling and neuronal migration factor (NSMF), and stromal derived growth factor 1 (SDF-1). Available data indicate that SOX10 promotes development of OECs.

Correct targeting and movement of GnRH neurons depends on multiple cues provided by many factors. These signals may act directly or indirectly through the scaffold of olfactory neurons. Chemokine gradients likely influence GnRH neuronal movement. Such factors include SDF-1 and gamma-aminobutyric acid (GABA). Curiously, SDF-1 and GABA appear to exert divergent effects to accelerate or retard, respectively, neuronal migration. SDF-1 acts through its receptor, CXCR4, via a G protein-activated inward rectifier potassium channel. SDF-1 has been observed in the nasal mesenchyme (NM), whereas its receptor, CXCR4, has been localized in migrating GnRH neurons and olfactory/vomeronasal nerve axons. Further, CXCR4-deficient mice exhibit a loss of GnRH neurons and impaired migration, suggesting the importance of SDF-1/CXCR4 signaling in the development of this system.

CCDC141 encodes a coiled-coil domain containing protein that is expressed in GnRH neurons and olfactory fibers. Knockdown of Ccdc141 did not change olfactory axon outgrowth, but was associated with decreased GnRH cell migration out of the nasal pit. In mice, Ccdc141 expression, correlated with migration in nasal regions and decreased when GnRH neurons entered the forebrain, appears to affect cellular motility through its interactions with myosin II.

Additional adhesion and guidance molecules include ANOS1, ephrins, and prokineticin 2 (PROK2). The gene encoding ANOS1 (previously known as KAL1 ) is located at Xp22.3 in the pseudoautosomal region of the X chromosome. Anosmin-1 is an extracellular matrix glycoprotein that contains a whey acidic protein-like protease inhibitor domain and four fibronectin type III domains. It promotes the formation of the lateral olfactory tract and neurite development. Anosmin may also serve as (1) an adhesion molecule to guide migrating GnRH neurons; and (2) as a chemoattractant for olfactory axon pathfinding. Also, it may interact with FGFR1. Ephrins are cell surface molecules that play a major role in axon guidance and signal through their cognate membrane tyrosine kinase receptors. PROK2 signals through the prokinecticin receptor 2 (PROKR2), a member of the rhodopsin G protein-coupled receptor family. PROK2 and its receptor (PROKR2) , a G protein-coupled receptor, appear to play major roles in olfactory bulb neurogenesis and GnRH neuron migration. However, neither protein is expressed in GnRH neurons. Based on the findings in mice with targeted PROK2 mutations, the GnRH neurons appear to be trapped with olfactory neurons with arrested migration just after crossoing the cribriform plate.

Semaphorins comprise a large and diverse family of secreted and membrane-associated proteins that influence the navigation of growing axons, and play a role in neural network formation. Four class 3 semaphorins, Sema3A, Sema3B, Sema3C, and Sema3F, are expressed around the developing olfactory/vomeronasal region. Semaphorin-3A (SEMA3A) is a secreted protein with repulsive effects on primary olfactory axons expressing the coreceptor neuropilin-1 (Nrp1), which may influence the migration of GnRH neurons. Semaphorin 3A is also expressed byOECs. Semaphorin-3E (SEMA3E) on the other hand protects maturing GnRH neurons from cell death. Semaphorin 4D is a membrane-bound semaphorin that can also be proteolytically released into the extracellular space in an active form. It can act as a proangiogenic factor through the coupling of its cognate receptor, PlexinB1, with the hepatocyte growth factor (HGF) receptor Met tyrosine kinase (MET). Both semaphorin 4D and PlexinB1 are highly expressed in the developing olfactory placode and the developing NM.

Semaphorin 7A (Sema 7A) appears to play a role in GnRH neuronal migration. Two mechanisms have been described: it can act as a membrane-bound signaling molecule or, following proteolytic cleavage, as a soluble factor. Sema7A can interact with two different receptors, plexin C1 and β1-integrin. Binding to plexin C1 decreases integrin-mediated cell attachment and spreading and interacting with β1-integrin induces integrin clustering and the activation of MAPK pathways. The phenotype of the mouse model with GnRH neuron-specific β1-integrin conditional KO showed impaired migration of GnRH neurons, delayed pubertal onset, and impaired fertility in female mice. In addition to its role in the development of the GnRH system, Sema7A appears to mediate the plasticity of GnRH neurons and tanycytes in the adult median eminence.

IGSF10, a member of the immunoglobulin superfamily, is also implicated in GnRH neuronal migration. Tissue expression studies using mouse embryos showed IGSF10 mRNA expression was localized to embryonic NM during the time that GnRH neurons are migrating through the NM. In a zebrafish IGSF10 knockdown model, loss of IGSF10 led to perturbed migration and failed neurite extension of GnRH3 neurons toward the hypothalamus.

HGF, Axl, and Tyro3 maintain GnRH neuronal survival when the neurons are crossing the cribriform plate region. HGF signals through its receptor, cMet, to promote GnRH neuron migration. Axl and Tyro3 are members of the TAM family of tyrosine kinase receptors and contain a fibronectin domain that binds to heparan sulfate proteoglycans. Mice with targeted Axl/Tyro3 mutations show impaired sex hormone-induced gonadotropin surge, resulting in estrous cycle abnormalities. The protein, growth arrest specific 6 (Gas6) encoded by Gas6, is a ligand that activates Axl and Tyro3. Gas6 is a heparan sulfate proteoglycan activated ligand with similarities to the FGFs and HGF. The phenotype of Gas6 knockout mice is characterized by early loss of GnRH neurons during embryonic development. Despite persistent decrease in GnRH neurons and impaired early stages of sexual maturation, these mice eventually manifested normal fertility.

FEZF1 is a zinc-finger gene encoding a transcriptional repressor that is highly and selectively present during embryogenesis in the olfactory epithelium. Fezf1 -deficient mice have impaired axonal projection of pioneer olfactory receptor neurons that cross the cribriform plate and subsequently innervate the olfactory bulb. These mice have smaller olfactory bulbs and an absence of GnRH neurons in the brain. Thus it appears that the FEZF1 product is required for the olfactory receptor neurons, and hence accompanying GnRH neurons, to enter the brain.

The roles of microRNAs (miRs) in neuronal development and maturation are becoming apparent. Data obtained using mice with targeted deletion of Distal-less-related 5 (Dlx5) gene demonstrated that specific miRs (-9 and -200 class) influence olfactory and GnRH neuron development. Mice with gonadotrope specific deletion of Dicer exhibited suppressed gonadotropin β-subunits and infertility. Another microRNA, miR-7a2, is expressed in pituitary gonadotropes. The phenotype associated with genetic deletion of miR-7a2 in mice includes low gonadotropin concentrations and infertility. miR-7a2 is highly expressed in the pituitary, but does not appear to influence GnRH neuron migration.

Upon arrival in the hypothalamus, the GnRH neurons extend projections to the median eminence to form a network that can secrete GnRH into the primary plexus of the hypophysial portal circulation. LH and FSH reach detectable levels by the 10th week of gestation in the human pituitary, peak in midgestation, and are higher in female fetuses than male fetuses. Although the hypothalamic control of the fetal pituitary gonadal axis has not been extensively studied in higher primates, the GnRH pulse generator is clearly driving the gonadotrophs of the fetal pituitary around the 15th week of gestation. Functional activity of this hypothalamic-pituitary system is essential for fetal testicular testosterone synthesis by the Leydig cell and normal male sex development. In contrast to the fetal testis, the ovary at this stage of development is relatively quiescent, and the absence of gonadal feedback signals likely accounts for the higher gonadotropin levels in the female fetus. As gestation progresses, the secretion of estradiol and other steroids by the fetoplacental unit increases dramatically, and suppresses gonadotropin secretion from the fetal pituitary by exerting an inhibitory action either directly at the pituitary or indirectly on the hypothalamus to restrain GnRH release.

Postnatal Development of Gonadotropin-Releasing Hormone Pulsatility

Following birth, GnRH pulse generator activity is robustly expressed, presumably due to the loss of placental steroids, and the pituitary gonadotrophs of the infant respond with LH and FSH secretion. At the hypothalamic level, kisspeptin and NKB are found in the arcuate nucleus at this stage of development, and a loss-of-function mutation in KISS1R has been associated with HH during infancy. Moreover, in the infant boy, the Leydig cells of the testis are stimulated so that circulating testosterone levels are similar to those observed in adult men. Peak testosterone concentrations occur at approximately 2 to 3 months of age and typically decline by 6 months of age. Among preterm male infants, LH and testosterone concentrations are higher than among full-term infants; phallic growth was positively correlated with urinary testosterone levels, and testicular growth was positively correlated with urinary FSH levels. Despite the “adult-like” endocrine milieu of increased gonadotropin secretion and elevated testosterone concentrations during the first few months of life, sexual hair does not develop and gametogenesis is not initiated, presumably due to limited AR signaling in skin and the immature Sertoli cell. The hypothalamic-pituitary-ovarian axis is active in term and preterm girls, and is associated with a transient increase in antral follicles.

Full hormonal responsivity of the gonad is acquired during childhood, but by this stage of development the GnRH pulse generator has been brought into check, resulting in the hypogonadotropic state that guarantees continued gonadal quiescence until the prepubertal phase of development is terminated by a resurgence of GnRH pulse generator activity ( Fig. 17.6 ). During childhood and juvenile development neither the GnRH neurons, pituitary gonadotrophs, nor the cells of the gonads are limiting to the onset of gonadarche.

A schematic of the pattern of pulsatile gonadotropin-releasing hormone (GnRH) release during juvenile development (including childhood) in boys (above) and girls (below) .

Insets indicate frequency of pulsatile GnRH release at respective stages of development.

(From Plant TM: Control of onset of puberty in primates. Topic Endocrinol 20:1, 2002.)

Accordingly, during these stages of development when the primate hypothalamus or pituitary is provided with pulsatile neurochemical, for example, N-methyl-D-aspartate (NMDA), a glutamate receptor agonist, or GnRH stimulation, respectively; or, when the ovary and testis are stimulated directly with LH and FSH, gonadarche may be readily elicited. Blocking endogenous GABA inhibition with the GABA(A) receptor blocker, bicuculline, dramatically increases kisspeptin release, implicating this neuropeptide in the onset of puberty. GABA plays a role in GnRH neuron migration and the onset of puberty. However, the precise details of GABA actions remain to be clarified. These manipulations that trigger premature GnRH secretion are graphically illustrated in children with GnRH-dependent precocious puberty, a disorder discussed later in this chapter.

That the GnRH neuronal network in the juvenile monkey hypothalamus is able upon NMDA stimulation to immediately elicit an adult hypophysiotropic GnRH drive is consistent with the finding that hypothalamic GnRH gene expression and peptide content are maintained throughout this phase of prepubertal development. Thus transcriptional regulation of the gene encoding GnRH from birth to puberty appears to be minimal, and the locus of the developmental control of GnRH release must lie upstream to the GnRH neuronal network.

Because GnRH is secreted in only picogram quantities into the hypophysial portal circulation, changes in the concentration of this neuropeptide in the peripheral circulation do not reflect hypothalamic activity. Therefore studies of the dynamics of the pubertal resurgence of GnRH pulse generator activity in man and other higher primates have generally utilized the high-fidelity relationship that exists between the frequencies of pulsatile GnRH release and episodic LH secretion. The latter may be tracked with relative ease by measuring moment-to-moment changes in LH concentrations in the peripheral circulation. Although the pubertal increase in hypothalamic GnRH drive to the gonadotroph probably involves both frequency and amplitude modulation of the GnRH pulse generator, the relationship between GnRH and LH pulse amplitude is more complex than the relationship between frequency because amplitude modulation of LH release may not always reflect changes in GnRH pulse amplitude. During the initiation of gonadarche in both boys and girls, LH pulse frequency accelerates and LH pulse amplitude increases in association with an amplification of a preexisting sleep-related diurnal pattern in release. This change in neuroendocrine activity may occur before the physical changes of gonadarche are manifest. In boys in particular, LH pulse frequency appears to decline later in pubertal development, probably due to a negative feedback action of rising testosterone concentrations. A longitudinal study of the agonadal monkey suggests that, as in humans, the pubertal acceleration of pulsatile GnRH release is an early neurobiological event in the initiation of gonadarche, and that it is a rapidly completed process. Thus the slow tempo of the overall progression of puberty probably results from mechanisms downstream from the hypothalamus, and particularly at the level of the pituitary.

The control system that dictates the up-down-up pattern of GnRH pulse generator activity from birth until puberty may be viewed as a neurobiological “brake” (or central restraint, as it has been previously described in the pediatric literature) that holds GnRH neuronal activity in check during the greater part of prepubertal development. Here it is important to recognize that the notion of a brake is conceptual; that is, the pubertal resurgence in robust GnRH pulsatility could be occasioned either from the removal of an inhibitory input or by the application of a stimulatory signal to the GnRH pulse generator, or a combination of the two. A similar argument may be applied to the earlier transition between infancy and childhood when GnRH pulsatility is markedly diminished. The neurobiological brake on pulsatile GnRH release throughout childhood and juvenile development is imposed in the absence of the ovary or testis. Consequently, the characteristic pattern of gonadotropin secretion observed during postnatal development in humans with robust gonadotropin secretion during infancy and puberty—separated by a prolonged hiatus in LH and FSH secretion—is maintained in the agonadal situation ( Fig. 17.7 ).

The neurobiological restraint that holds the gonadotropin-releasing hormone pulse generator in check during childhood is imposed in the absence of the gonads, as reflected by the time courses of circulating concentrations of luteinizing hormone (LH; top) and follicle-stimulating hormone (FSH; bottom) in 58 patients with gonadal dysgenesis.

Closed triangles , 45,X karyotype; circles , X chromosome mosaicism or structural abnormalities; cross-hatching, mean plasma gonadotropin concentrations in normal girls. Bold curves represent polynomial regressions of gonadotropin concentration with age.

(From Conte FA, Crumbach MM, Kaplan SL: A diphasic pattern of gonadotropin secretion in patients with the syndrome of gonadal dysgenesis. J Clin Endocrinol Metab 40:670, 1975.)

Similarly, in male infants with partial androgen insensitivity, LH concentrations are generally elevated, and this is associated with higher testosterone levels. Yet, infants with complete androgen insensitivity often fail to demonstrate a postnatal rise in LH and testosterone secretion. The later finding is counterintuitive, and an understanding of the molecular basis of this phenomenon may well reveal fundamental insights into the ontogeny of GnRH pulse generation. In agonadal children, the degree of the prepubertal suppression of gonadotropin release is less than that observed in eugonadal individuals. Interestingly, in agonadal children circulating gonadotropin levels are higher in girls than boys, indicating that the intensity of the neurobiological brake imposed on the GnRH pulse generator during prepubertal development is less in the female than in the male. As a result, the gonadotropin drive to the prepubertal ovary stimulates a low level of estradiol secretion, which, through negative feedback action on LH and FSH release, amplifies the relatively weaker neurobiological brake restraining gonadotropin secretion in the prepubertal girl. This sex difference in the strength of the neurobiological brake on prepubertal GnRH release is associated with a shorter duration of the brake in girls, which probably accounts for the relatively earlier age of gonadarche in the female. These and other sex differences in the developmental control of the GnRH pulse generator are presumed to result from greater exposure of the fetal male hypothalamus to testosterone.

Nature of the Neurobiological Brake

The discovery in 2003 that loss-of-function mutations in KISS1R in humans were associated with HH and delayed or absent puberty demonstrated the critical role of kisspeptin in regulating GnRH secretion. Subsequent studies using several different experimental models validated this finding. These data led to the proposal that a major component of the neurobiological brake imposed upon pulsatile GnRH release during the greater part of prepubertal development is due to a hiatus in a stimulatory kisspeptin input to the GnRH neuronal network. This proposal was based on findings in the monkey, that hypothalamic expression of KISS1 and release of kisspeptin in the region of the median eminence increase at the time of the pubertal resurgence in GnRH pulsatility. Additionally, intermittent administration of kisspeptin at hourly intervals during juvenile development elicits a precocious and sustained adult-like pulsatile pattern of GnRH, and the pubertal increase in GnRH release may be suppressed by the administration of a KISS1R receptor antagonist directly to the median eminence.

The finding in humans that loss-of-function mutations in the NKB signaling pathway are associated with a phenotype similar to that reported earlier for inactivating mutations in KISS1R, together with the observation that these two neuropeptides are coexpressed in the same neurons (KNDy neurons) in the arcuate nucleus, have led to the concept that these KNDy neurons are responsible for the generation of GnRH pulsatility. Thus kisspeptin expressing KNDy neurons in the arcuate nucleus comprise a critical component of the GnRH pulse generator. In the absence of an intact kisspeptin signaling pathway, the output of the GnRH pulse generator will be abrogated and pulsatile GnRH release will be compromised, resulting in a delay or absence of puberty. Overall, available data suggest that the KNDy neurons in the arcuate nucleus themselves do not govern the timing of puberty; rather, these neurons appear to be subservient to upstream regulatory mechanisms that govern the developmental pattern of pulsatile GnRH release and the onset of puberty ( Fig. 17.8 ).

A model for the control of the timing of puberty.

The role of kisspeptin signaling is posited to be a critical component of the neural machinery that generates pulsatile gonadotropin-releasing hormone (GnRH) release from the hypothalamus (the hypothalamic GnRH pulse generator). In this model, the hypothalamic GnRH pulse generator resides in the arcuate nucleus (ARC) and the output of the pulse generator is relayed to GnRH terminals (red) in the median eminence (ME) by kisspeptin projections arising from KNDy neurons in the ARC (green) . During infancy (left panel), GnRH pulse generator activity is robust, leading to intermittent release of kisspeptin in the ME, resulting in a corresponding pattern of GnRH release into the portal circulation. This, in turn, drives pulsatile gonadotropin secretion. In the transition from infancy to the juvenile phase of development (middle panel), a neurobiological brake (central inhibition) restrains the GnRH pulse generator, and pulsatile release of kisspeptin in the ME is markedly suppressed. This leads to reduced GnRH release and to a hypogonadotropic state in the juvenile period. Puberty is triggered when the neural brake is released and GnRH pulse generator activity with robust intermittent release of kisspeptin in the ME is reactivated (right panel) . According to this model, the mystery of primate puberty lies in the nature of the neurobiological brake and in the mechanism that times its application during infancy and its release at the end of the juvenile phase of development. It should be noted that the ability of the postnatal gonad to respond fully to gonadotropin stimulation is not acquired until the juvenile stage of development, by which time luteinizing hormone (LH) and follicle-stimulating hormone (FSH) secretion is low as a result of the GnRH pulse generator being brought into check. The thickness of the blue (T, testosterone) and gold (E, estradiol) arrows indicating negative feedback by the testis and ovary, respectively, reflects the degree of gonadal steroid inhibition exerted on LH secretion at these three stages of development. AC , Anterior commissure; AP , anterior pituitary gland; MMB , mammillary body; OC , optic chiasm.

(From Terasawa E, Guerrier KA, Plant TM: Kisspeptin and puberty in mammals. In Kauffman AS, Smith JT, editors: Kisspeptin Signaling in Reproductive Biology , Springer, 2013, New York, p 253.)

The nature of the upstream pathways that comprise the neurobiological brake on the GnRH pulse generator during childhood and juvenile development remains poorly understood.

Studies of the female rhesus monkey provide evidence that GABA, the major inhibitory neurotransmitter in the brain, is upregulated during juvenile development, and inhibition of GABA tone in the hypothalamus of the prepubertal monkey leads to precocious menarche and ovulation. Interestingly, infusion of the GABA antagonist, bicuculline, into the median eminence of prepubertal female monkeys stimulates release of kisspeptin-54 into this region of the hypothalamus in association with that of GnRH, and the bicuculline-induced GnRH release is blocked by simultaneous infusion of a kisspeptin antagonist. However, it is unclear what reduces GABA inhibition prior to puberty, where the relevant GABAergic neurons are physically located, and how GABA signaling interacts with the GnRH pulse generator.

Other transynaptic signals implicated in the regulation of the pubertal resurgence of GnRH pulse generator activity include glutamate and neuropeptide Y (NPY). Glutamate is the major excitatory neurotransmitter in the brain and, in contrast to GABA, hypothalamic release of this amino acid is increased at the time of puberty in the female monkey, and as discussed earlier in the chapter, repetitive activation of glutamate receptors in the juvenile monkey rapidly leads to the onset of precocious gonadarche. NPY neurons are found in the arcuate nucleus and, in the male rhesus monkey, NPY gene expression in the hypothalamus is inversely related to the up-down-up pattern of GnRH pulse generator activity from birth to puberty. NPY receptors are inhibitory G protein receptors, and their activation leads to hyperpolarization and inhibition of neural activity. However, pharmacological approaches failed to demonstrate that inhibition of NPY signaling in the hypothalamus of the juvenile monkey did not promote GnRH release.

While neuroglia have classically been regarded as subserving only a “supporting role” in the central nervous system (CNS), contemporary views hold that these non-neuronal cells play important functional roles within the brain. Moreover, in the context of the hypothalamus, secretion of TGF-α by astroglia has been postulated to provide the GnRH neuronal network with a stimulatory input at the time of puberty.

Attempts to elucidate the neural mechanism dictating the postnatal pattern of GnRH pulse generation have traditionally led investigators to focus in relative isolation on a “favorite” signaling pathway. Using a systems biology approach, global gene discovery has been combined with computational (in silico) biology to identify functional linked networks of hypothalamic genes that are found to be associated with changes in GnRH pulse generator activity. The initial gene discovery approach was conducted without regard to the phenotype of the cells in which the respective genes are expressed and gene networks are operating. Available data indicate that the developmental changes in the transcriptional factors and gene network relevant to GnRH secretion lie upstream of the KISS1 gene and are, therefore, upstream of GnRH pulse generation. This network of genes serves as a governing hierarchy to orchestrate the resurgence of pubertal GnRH release and, therefore, modulate the timing of puberty. Since such networks of genes are further proposed to function in the absence of signals derived from the periphery, conceptually they may be viewed at a systems level as comprising a pubertal clock (discussed later in the chapter).

Expression of two such transcriptional regulators, enhanced at puberty 1 ( EAP1, also known as interferon regulatory factor 2 binding protein-like) and thyroid transcription factor-1 ( TTF-1, also known as NKX2-1 ), increase in the mediobasal hypothalamus of nonhuman primates at puberty. EAP1 is expressed in kisspeptin neurons in the arcuate nucleus of the monkey. Moreover, its expression in the hypothalamus increases at the time of puberty in the female monkey, and the knock down of EAP1 using a lentivirus approach interrupts menstrual cyclicity in the adult female. Additionally a single nucleotide polymorphism (SNP) upstream of the EAP1 gene has been associated with irregular menses in the monkey. Conditional deletion of Ttf1 from terminally differentiated hypothalamic neurons was associated with delayed puberty, decreased Kiss1 expression, and subfertility. Genetic analysis failed to identify germline mutations in either EAP1 or TTF-1 in patients with HH. Nevertheless, both EAP1 and TTF-1 are functionally connected to genes identified by genome-wide association studies (GWAS) to influence age at menarche.

Additional evidence supporting the concept of a network of genes determining the level of function of the GnRH pulse generator have identified a cohort of genes that encode for a group of transcriptional suppressor proteins known as the polycomb group. In the pubertal rat, expression of these genes is downregulated by DNA methylation, leading to a reduction in the silencer proteins. Two of the polycomb group genes are expressed in the arcuate nucleus, and overexpression of one of these genes resulted in a decrease in Kiss1 expression in association with delayed vaginal opening and a disruption of GnRH pulsatility in mediobasal hypothalamic (MBH) explants. A global inhibition of DNA methylation in prepubertal rats resulted in the delay of vaginal opening, indicating that epigenetic regulation of gene expression may be important for timing puberty. Similarly, data obtained in monkeys indicate that some zinc-finger transcriptional repressors restrain puberty by epigenetically repressing a gene network that operating in the arcuate nucleus controls puberty by governing pulsatile GnRH release.

Additional support for a zinc-finger motif containing suppressor gene holding puberty in check has come from human mutation studies. Using whole exome sequencing (WES), mutations in the gene encoding for makorin RING finger protein 3, MKRN3 , are associated with central precocious puberty (CPP) in seven girls and five boys from 12 families. All affected individuals inherited the disorder-associated allele from their fathers, indicating that only the paternal allele is expressed. These mutations are predicted to result in loss of function of the protein. While the function of MKRN3 is not fully understood, the makorin family of proteins contains a particular zinc-finger motif that has been associated with ubiquitination, a process that is involved in protein trafficking, which in some cases leads to protein degradation. Relevance of this gene to puberty is further supported by demonstrating declining levels of circulating MKRN3 product prior to pubertal onset in Danish girls and boys. Expression of this gene decreases immediately before the onset of puberty in the mouse hypothalamic arcuate nucleus. Thus in mice, makorin-3 may contribute to the neurobiological brake on GnRH secretion. Together, these studies support the concept that transcriptional repression is a core component of the neuroendocrine circuitry that regulates the timing of puberty.

Recently, genome-wide association studies indicate that several specific loci on the human genome are associated with variations in the age at menarche. An SNP that was consistently and strongly associated with an earlier age at menarche was found on chromosome 6 near LIN28B, which encodes for a micro-RNA binding protein. The differences in age at menarche in subjects with and without variants in the region of LIN28B are small (1 to 2 months) compared to the recognized range for the age at menarche in the population at large that may extend from 10 to 16 years of age. A genetic study on girls with early puberty did not find any potentially responsible LIN28B mutations. It is unclear at the moment the exact causal role of LIN28-Let7 if any in timing of human puberty. A follow-up meta-analysis study using genome-wide and custom-genotyping arrays in more than 180,000 European women found strong evidence for 123 SNPs at 106 gene loci to be associated with earlier age at menarche. Surprisingly, only a few known puberty genes overlap with the genes associated with age at menarche. These include MKRN3 , LEPR , IGSF1 , and TACR3 , further implicating biological significance of these genes in pubertal development.

Regardless of the components of the neurobiological brake, which dictate the up-down-up pattern of pulsatile GnRH release from birth until puberty, it is to be anticipated that application and withdrawal of the brake will be associated with a corresponding structural remodeling in those hypothalamic neuronal and glial circuits involved. In this regard, the hypothalamus of the postnatal brain retains its capacity for plasticity, as reflected by the expression of polysialic acid-neural cell adhesion molecule (NCAM).

Putative Physiological Control Systems Governing the Timing of Gonadarche

The physiological control systems that dictate the timing of puberty have intrigued investigators for decades, but the governance of this fundamental developmental event in humans remains largely a mystery. However, two basic schemata have been proposed. In the first, a cue to reawaken the GnRH pulse generator is provided by the attainment of a particular state of somatic maturation. According to this hypothesis, the brain receives this information by way of a signal in the circulation that is tracked by a somatometer resident within the CNS. In the second schema, a pubertal clock (presumably resident in the CNS) generates the signal.

In the case of the somatometer hypothesis, the attainment of a particular proportion of body fat has long been argued to be necessary for the onset of gonadarche. Interest in this hypothesis was rekindled with the discovery of leptin and its receptor. Leptin, encoded by the leptin (LEP) gene, is primarily secreted by adipocytes and regulates feeding behavior and body weight by providing the hypothalamus with information on fat mass and energy status. The leptin receptor (LEPR) is a single transmembrane type I cytokine receptor of the IL-6 receptor family that is encoded by the LEPR gene located at chromosome 1p31.3. Although multiple LEPR isoforms exist, leptin action is primarily mediated by the long form of the LEPR.

Individuals with mutations in leptin signaling, resulting from loss-of-function mutations in either LEP or LEPR, have failed to progress through puberty. The phenotype of LEPR mutations included morbid obesity, abnormal eating behaviors, and lack of spontaneous pubertal development. Over the last 15 years, leptin replacement has been administered to several patients with leptin deficiency due to LEP mutations. When administered at an age appropriate for puberty, leptin promoted pubertal development. Importantly, there has been no evidence of premature puberty in younger children following replacement treatment. Moreover, among children with GnRH-dependent precocious puberty, leptin concentrations correlated with BMI and not pubertal status. These findings provide compelling evidence that the action of leptin, albeit obligatory for the onset of puberty, is nevertheless permissive. This action of leptin may require only low circulating levels of the adipocyte hormone because pubertal development in both male and female subjects with various lipodystrophies has been reported to be normal despite low leptin concentrations.

In girls, it is generally recognized that plasma leptin levels increase progressively through breast stages 1 to 5. In boys, plasma leptin levels also increase during early puberty to reach peak levels between 10 and 12 years of age, but subsequently decline as blood levels of testosterone rise into the adult range. The finding that circulating concentrations of the soluble form of the LEPR progressively decrease during childhood until approximately 11 years of age suggests that the increase in bioavailable leptin during early puberty may be greater than that reflected in total leptin concentrations.

Considerable interest is focused on identification of leptin’s neuronal targets that promote pubertal progression. While it is generally recognized that GnRH neurons do not express LEPR , data obtained in rodents and sheep for the KNDy neurons are inconsistent and the site of action of leptin may be upstream to the GnRH pulse generator. One potential mechanism demonstrated in mice is that leptin induces phosphorylation of neuronal nitric oxide synthase (nNOS) in nNOS-expressing neurons in the preoptic region, which leads to increased LH secretion independent of kisspeptin/KISS1R signaling. Thus NO signaling may play a major role in the crosstalk between leptin and the reproductive axis.

Before leaving the subject of leptin and obesity, it should be noted that although overweight girls tend to “mature” earlier, sleep-related increments in LH secretion, and therefore presumable GnRH pulse generator activity, have recently been reported to be blunted in healthy obese premenarcheal girls.

Other somatic factors have been proposed to serve as signals to the somatometer. The hypothesis that such a factor is skeletal in origin is based on the finding that, in children with an accelerated or retarded maturational tempo, menarche and testicular enlargement correlate better with skeletal age than with chronological age. Skeletal age, also known as “bone age,” is a surrogate marker of biological maturation and may be determined by comparing a radiograph of the left hand to gender-specific standards obtained at various chronological ages. Several caveats are relevant when assessing skeletal maturity. The degree of skeletal maturation at various sites, such as hands, elbows, and knees, may differ. Skeletal maturation of the carpal bones, distal radius, and distal ulna often lag behind the metacarpals and phalanges. Nevertheless, the association between bone age and the onset of gonadarche is maintained in disorders of growth. In children with constitutional delay of growth and with true isolated growth hormone (GH) deficiency, gonadarche occurs at a late chronological age but at a normal skeletal age. On the other hand, when skeletal maturation is advanced, as may occur in association with CAH or familial testotoxicosis, secondary GnRH-dependent precocious puberty may develop. Although proteins synthesized in bone enter the vascular compartment, the ability of osteocalcin and other bone proteins to modulate the activity of the GnRH pulse generator has not been addressed.

GH secretion during childhood is relatively stable, but GH release is amplified up to threefold in boys and girls with the initiation of gonadarche and the rise in circulating levels of sex steroids. The pubertal increase in GH, in combination with the increased circulating concentrations of insulin-like growth factor-1 (IGF-I), estrogens, and androgens, contribute to the adolescent growth spurt. The increased secretion of GH at the time of gonadarche is not sustained; by late puberty, GH levels begin to decline. During puberty, height velocity, IGF-1 concentrations, and sex steroid concentrations rise with synchronization of these changes within individuals. Because the increases in GH and IGF-I appear to be in response to the initiation of gonadarche, and particularly to increased gonadal steroid secretion at this time, neither GH nor IGF-I represent compelling candidates for the signal responsible for the resurgence of GnRH release.

At the time of gonadarche, insulin resistance increases. This insulin resistance is greatest among children in Tanner stages 2 and 3 when compared with prepubertal children and adults. Manifestations of insulin resistance appear to be limited to effects on carbohydrate metabolism and are associated with compensatory hyperinsulinemia and normal disposition index. Disposition index, a function of insulin sensitivity and insulin secretion, reflects beta-cell response for a given insulin sensitivity. Using euglycemic-hyperinsulinemic clamp studies in conjunction with investigation of substrate utilization, one longitudinal study found that puberty was associated with decreased insulin sensitivity, increased insulin secretion, increased total body lipolysis, decreased glucose oxidation, and increased IGF-1 concentrations. In a longitudinal prospective cohort study involving healthy children, insulin sensitivity decreased before the onset of physical features of puberty. In one study, no changes in insulin’s ability to suppress hepatic glucose production were noted during puberty, indicating that decreased insulin sensitivity is limited to peripheral glucose uptake. Although cross-sectional studies suggest that the magnitude of insulin resistance is influenced by BMI (body weight/height [kg/m ² ]), gender, and ethnic background, these associations have not been consistently noted in longitudinal studies. IGF-1 concentrations during puberty have been reported to mirror those in insulin sensitivity. Mean 24-hour serum GH and IGF-I concentrations positively correlate with the degree of insulin resistance during puberty, and the pubertal changes in insulin sensitivity may be partially mediated by increased GH and IGF-I concentrations. Yet, while it is generally recognized that increased gonadal steroid levels are responsible for activation of the GH/IGF1 axis at puberty (discussed earlier in the chapter), a causal relationship between testosterone or estradiol and insulin resistance has not been demonstrated.

Treatment of nonobese Catalunyan girls with premature adrenarche and advanced skeletal maturation with the insulin sensitzer, metformin, was associated with later onset of breast development and menarche. Moreover, in this relatively homogenous ethnic population, metformin treatment for 3 years slowed pubertal tempo among low birth weight girls who are at risk for early puberty and shorter adult stature. Taken together, the foregoing considerations raise the possibility that decreased insulin sensitivity may represent a component of the cue that times the onset of gonadarche.

Ghrelin is a small peptide, secreted predominantly by the stomach, which circulates in two forms. The active form is acetylated and promotes GH secretion. Ghrelin influences food intake, sleep, body weight, gastrointestinal mobility, and reproduction. Ghrelin suppresses LH pulsatility in the pituitary. Circulating ghrelin concentrations are higher during fasting, and decrease after food intake, indicating that ghrelin signals energy deficient states and modulates appetite and carbohydrate metabolism. Ghrelin concentrations peak during the first 2 years of life and decrease during puberty. Negative correlations were found between ghrelin concentrations and both age and pubertal stage. The ghrelin receptor, GH secretagogue receptor-1a (GHSR-1a), is expressed in human hypothalamus, pituitary, testis, and ovaries.

Ghrelin and leptin appear to act as reciprocal regulators of energy homeostasis exerting opposing influences on the HPG axis. Adipokines, hormones secreted by adipocytes, include resistin, adiponectin, leptin, and visfatin. Adiponectin concentrations are inversely related to insulin resistance and have been reported to decrease in pubertal males. Resistin is secreted by adipocytes and peripheral blood mononuclear cells and signals through the Toll-like receptor 4. Although details regarding its physiologic role are unclear, resistin appears to promote insulin resistance. The respective roles, if any, of resistin and visfatin on gonadarche or adrenarche remain to be determined.

Nutrition and Diet

Adequate energy stores are essential for the accelerated linear growth and achievement of reproductive competence at puberty. Neuropeptides and hormones provide information regarding energy status. Nutritional state modulates the onset of gonadarche in girls, as reflected by the findings that menarche is delayed in malnourished girls, and menarche tends to occur at a particular or “critical” body weight rather than at a set age. Obesity is associated with early breast development and menarche. Rapid weight gain during the first year of life is associated with obesity and early menarche in a prospective UK cohort study. In addition, the timing of adrenarche may also be influenced by nutritional status. In the female rhesus monkey, high calorie diets were associated with a precocious increase in nipple volume, sex-skin swelling, and menarche in association with increased concentrations of leptin and IGF-1 concentrations, and an increased BMI. While the impact of undernutrition on gonadarche in boys has received less attention, there is no reason to suspect that the male axis is unaffected in this regard.

Leptin and osteocalcin appear to communicate energy status to several tissues including adipose tissue, bone, CNS, and gonads. In its undercarboxylated (active) form, osteocalcin promotes insulin secretion, improves insulin sensitivity, induces β-cell proliferation, increases adiponectin secretion, and decreases lipolysis. In osteoblasts, insulin signaling increases bone formation, bone resorption, and secretion of active osteocalcin production.

Differences in the timing and tempo of gonadarche unrelated to racial or ethnic background have been reported even among well-nourished girls. Researchers have suggested that diet may account for such variations. Attention has been drawn to subtle relationships between diets high in animal protein and early menarche, and between diets high in vegetable grains and delayed onset or tempo of gonadarche. Moderate to vigorous exercise in the absence of weight restriction appears to have a negligible impact on the timing and tempo of puberty in either sex. However, breast development, menarche, and skeletal maturation can be delayed in girls involved in strenuous physical training. The association between extreme energy expenditure and delayed gonadarche in the female is particularly marked in ballerinas, long-distance runners, and figure skaters, who must maintain their body weight within strict limits. Such girls may experience amenorrhea secondary to HH. The factors responsible for compromising pulsatile GnRH release in pubertal children who exercise vigorously are presumed to be similar to those resulting from undernutrition. The life of the young female dancer or athlete is stressful; therefore stress may also represent a contributing factor underlying exercise-induced delayed gonadarche. Because certain physical and psychological characteristics are necessary for outstanding athletic performances, there may be a significant contribution of self-selection involved in the decision of girls to participate at such an exceptional level. This can be complicated by eating disorders, which are more common in elite artistic athletes such as gymnasts. In a study of young female gymnasts and their parents, menarche in the mothers was found to be delayed relative to that of mothers of sedentary girls. Therefore, the possibility of a genetic contribution to this phenomenon cannot be totally excluded. The effect of physical training on the timing of pubarche in girls has been less studied, and conclusive data are lacking.

Although the relationship between strenuous physical training and male puberty has received less attention, this developmental process is apparently less susceptible in boys than it is in girls. This may be related, in part, to sex differences in the age at which training is initiated, and to the intensity of the exercise. Nevertheless, in sports like wrestling that require weight control achieved with a combination of strenuous exercise and dietary restriction, impaired testosterone secretion may occur.

Environmental Disruptors

Increased attention has been focused on EDCs as modulators of pubertal timing. Most EDCs have estrogenic or anti-androgenic activities. Many occur as mixtures rather than single agents and may have multiple actions. Sexually dimorphic actions can occur. Dioxins, PCB, PBB, BPA, plasticizers (phthalates), pesticides, fungicides (vinclozolin), alcohol, tobacco, and pharmaceutical agents (diethylstilbestrol) are considered to be EDCs. Phytoestrogens, which can function as selective estrogen receptor modulators (SERMS), are found in soybeans, flaxseed, peanuts, and some vegetables. Most are diphenolic compounds with structural features common to estrogenic steroid agonists and antagonists. EDCs can affect peripheral reproductive systems, such as inducing breast development. These chemicals may also exert neuroendocrine effects to influence hypothalamic and pituitary function. EDCs can act through hormone receptors, can affect enzymatic processes, and may disrupt the complex interactions of endogenous hormones at multiple levels. Available data suggest that EDCs are associated with both early puberty and delayed completion of puberty. The specific consequences of EDCs may depend on age and duration of exposure. Further investigations into EDCs, defining mechanisms of action and consequences, are necessary.

Prenatal Influences

The roles of prenatal influences and environment on postnatal development including puberty are increasingly recognized. Children with intrauterine growth retardation (IUGR) or born small for gestational age (SGA) have increased risks to develop insulin resistance, hypertension, diabetes mellitus, metabolic syndrome, and coronary artery disease in adulthood. Increased weight gain during the first few years of life exaggerates these risks. Increased adrenal androgen levels and in some cases precocious or exaggerated adrenarche has been reported in both boys and girls born SGA. Some, but not all, girls with premature adrenarche have an increased risk to develop polycystic ovary syndrome (PCOS) during adolescence. Reports are inconsistent regarding the timing of menarche in girls born SGA compared to girls born at an appropriate size for gestational age. The timing of puberty in boys born SGA appears to be normal, although subfertility has been reported after they reached adulthood. Perinatal HIV infection is associated with later onset of puberty. Potential mechanisms responsible for the consequences of abnormal fetal growth are multifactoral and not mutually exclusive. Suggested causes include genetic variations, differential methylation, alterations in microRNA, altered histone acetylation, and other long-range regulatory effectors.

The neuroendocrine consequences of prenatal androgen exposure have been best characterized in sheep. Findings include decreased sensitivity to estradiol and progesterone feedback, decreased content of dynorphin and NKB in KNDy neurons, and altered morphology of KNDy neurons. Whether these data are applicable to humans and whether prenatal androgen exposure alters GnRH pulse generator function is unclear.

Adoption or Migration from Developing to Developed Countries

Several European countries have reported precocious gonadarche in a relatively dramatic proportion of children—particularly girls—adopted from developing world regions like Asia and South America. This increased endocrine activity in the hypothalamic-pituitary-ovarian axis prior to the onset of physical manifestations of puberty indicates a central origin to the condition. The cause of the precocity appears to be complex, but improved nutritional and social conditions may contribute to this phenomenon. Precocious gonadarche is also seen in immigrant girls arriving with their parents and lacking evidence of earlier compromised growth and nutrition. Moreover, ethnic background does not appear to influence this trend. A study of immigrant girls in Belgium with premature gonadarche reported an association with previous exposure to organochlorine pesticides. A study in Spain found an increased relative risk for precocious gonadarche in adopted girls (both national and international), but not in immigrant girls. However, this outcome is inconsistent because adopted Chinese girls experienced menarche at ages comparable to nonadopted Chinese girls.

Disorders of Early Puberty

Gonadotropin-Releasing Hormone-Dependent Precocious Pubertal Development

Progressive Precocious Gonadarche or Central Precocious Puberty

Although this disorder is labeled GnRH-dependent precocious puberty or puberty (central precocious puberty [CPP]), it represents precocious gonadarche due to either premature resurgence or incomplete suppression of the hypothalamic GnRH pulse generator. It occurs more often in girls than in boys. This sex difference is probably related to the lesser prepubertal suppression of the GnRH pulse generator in girls than boys. The sequence of pubertal development is typical of a normal puberty including adrenarche in some cases, but it begins at an earlier-than-normal age. Observational data collected from 2008 to 2010 through a stringent Spanish hospital-based registry reported an annual incidence ranging from 0.02 to 1.07 cases per million.

Approximately 90% of cases of CPP in girls are considered to be idiopathic whereas 50% to 70% of cases in boys are associated with identifiable etiology. The discovery of genes involved in GnRH and gonadotropin secretion has allowed for identification of specific gene mutations in patients with CPP. A heterozygous missense mutation, p.Arg386Pro, in the KISS1R gene was associated with precocious puberty in a girl. This mutation located in the C-terminal tail of the receptor was associated with delayed degradation of the receptor yielding a prolonged duration of action. In addition to this activating mutation in the receptor, a mutation in the KISS1 gene was identified in a young boy who presented at 17 months of age with CPP. This variant, p. Pro74Ser, appears to prolong the half-life due to resistance to degradation. To date, KISS1 and KISS1R mutations are extremely rare in children with CPP.

As described above, paternally inherited MKRN3 mutations are associated with CPP in boys and girls. This gene is mapped to chromosome 15q11.2 in the Prader-Willi Syndrome critical region. The maternal allele is methylated and silenced resulting in expression of the paternal allele. The gene encodes a protein involved in ubiquitination and cell signaling. Several unique mutations have been reported among different ethnic groups. Mutations in this gene appear to be the most common familial etiology of CPP. The median age of pubertal onset is 6.0 years (3.0 to 7.5) in girls and 8.25 years (5.9 to 9.0) in boys. No abnormalities of the CNS have been reported. Curiously, despite the physical proximity of MKRN3 to the Prader-Willi syndrome locus, no major signs of Prader-Willi syndrome have been described. A GWAS study looking at the parent of origin effects on puberty identified a signal in the region of MKRN3 gene; the paternal allele of a specific SNP (rs12148769, G>A) affects age at menarche in healthy girls suggesting that variants in this region affect pubertal timing within the normal range. Circulating makorin-3 concentrations were found to decline prior to puberty among healthy girls and boys ( Figs. 17.10 and 17.11 ).

Circulating MKRN3 levels decline prior to pubertal onset and through puberty: a longitudinal study of healthy girls.

Component variant model of serum concentrations of MKRN3 according to time to pubertal onset based on all longitudinal data. Black lines indicate geometric means (95% confidence intervals). The number of girls contributing with measurements to the different time points is listed above the x-axis. P -values correspond to analyses comparing reference levels (the last prepubertal examination) with levels at different time points. MKRN3 declined from 3 years before pubertal onset to the last prepubertal examination ( P = .033).

(From Hagen CP, Sørensen K, Mieritz MG, Johannsen TH, Almstrup K, Juul A: Circulating MKRN3 levels decline prior to pubertal onset and through puberty: a longitudinal study of healthy girls. J Clin Endocrinol Metab 100:1920–1926, 2015. Reproduced with the permission of the Journal of Clinical Endocrinology and Metabolism.)

Circulating MKRN3 levels decline during puberty in healthy boys.

Variance component model of serum MKRN3 levels according to time to pubertal onset/pubertal age based on longitudinal data. Present data of boys are marked as blue, whereas previously published data in girls are marked as red . The solid line indicates geometric mean (95% confidence interval, dotted line ). The number of boys contributing with measurements to the different time intervals is listed above the x-axis. P- values correspond to analyses comparing reference levels (the last prepubertal examination, −0.5-y interval, “REF”) with levels at different intervals in boys. Significant P- values are marked with asterisks. MKRN3 levels in boys declined from 5 years as well as 4 years before pubertal onset to the last prepubertal examination ( P < .001 and P = .01, respectively) and subsequently declined throughout puberty. Peripubertal MKRN3 levels differ substantially between healthy boys and girls.

(From Busch AS, Hagen CP, Almstrup K, Juul A: Circulating MKRN3 levels decline during puberty in healthy boys. J Clin Endocrinol Metab 2016;101:2588–2593. Reproduced with the permission of the Journal of Clinical Endocrinology and Metabolism.)

CPP has been described in the Williams-Beuren syndrome, histiocytosis X, and maternal uniparental disomy for chromosome 14. Evaluation of a family with CPP led to identification of a deletion/duplication mutation in the delta-like 1 homologue (DLK1) gene located at chromosome 14q32. The four affected individuals presented with thelarche between the ages of 6.4 to 8.0 years of age. An additional investigation revealed undetectable serum DLK1 concentrations (<0.4 ng/mL) compared to control group (1.9 to 20 ng/mL). DLK1 is a paternally expressed gene located within the genetic locus associated with Temple and Kagami-Ogata syndromes. Temple syndrome is characterized by IUGR, hypotonia in infancy, CPP, and short stature. Genetic findings in Temple syndrome include maternal uniparental disomy, paternal deletion, or loss of differential methylation at the DLK1/MEG3 region on chromosome 14. Patients with Kagami-Ogata syndrome is associated with prenatal overgrowth, developmental delay, abdominal wall defects, but disorders of pubertal timing do not appear to be a prominent feature. Based on available animal and human data, the opposing directions of regulation of MKRN3 and DLK1 expression prior to puberty may underlie interactions between these factors that modulate the timing of puberty.

Hypothalamic hamartomas, congenital malformations composed of a heterotropic gray matter, neurons, and glial cells usually located on the floor of the third ventricle or attached to the tuber cinereum, are a common etiology of CPP. Two potential mechanisms have been hypothesized for the association between CPP and hypothalamic hamartomas; one is increased GnRH secretion from tissue emancipated from suppression by the prepubertal brake and the other is that factors such as TGF-α provide an ectopic drive to GnRH neurons with a normal distribution in the hypothalamus. The hamartomas can be classified as parahypothalamic , attached or suspended from the floor of the third ventricle, or as intrahypothalamic , in which the mass is enveloped by the hypothalamus and distorts the third ventricle. The lesions do not grow over time, do not metastasize, and do not produce β-human chorionic gonadotropin(β-hCG) and α-fetoprotein. Extreme precocity suggests a hamartoma. Although gelastic or laughing seizures can be associated with precocious puberty due to hypothalamic hamartomas, the majority of patients with hypothalamic hamartomas do not exhibit neurological symptoms. Among patients in whom hamartomas were removed to treat intractable seizures, hamartomas associated with CPP were more likely to contact the infundibulum or tuber cinereum and were larger than hamartomas not associated with CPP. All hamartomas expressed GnRH, TGF-α, and GnRHR. No differences were detected in the expression of KISS1 and GPR54 between hamartomas associated with CPP and those not associated with CPP. Gene expression profiling of hypothalamic hamartomas associated with precocity may provide clues regarding genes, proteins, and regulatory pathways associated with the timing of puberty.

Hypothalamic hamartomas are generally sporadic, but may occur as a feature of several dysmorphic syndrome including Pallister-Hall syndrome and oral-facial-digital syndrome (OFD) types I and VI. Pallister-Hall syndrome is an autosomal-dominant disorder associated with mutations in the GLI3 gene located at chromosome 7p13. Additional features include pituitary anomalies, hypopituitarism, imperforate anus, and polydactyly. In the presence of sonic hedgehog (SHH), the full-length GLI3 translocates to the nucleus where it functions as a transcriptional activator. When SHH is absent, GLI3 is phosphorylated and cleaved; the cleaved protein translocates to the nucleus to repress specific target genes. The truncated GLI3 protein showed stronger repressor activity than the wild type protein, suggesting that these mutations are associated with constitutive repression of SHH signaling. Mutations in additional genes in the SHH pathway, especially PRKACA, have been detected in HH associated with epilepy.

In addition to hypothalamic hamartomas, optic gliomas, suprasellar cysts, arachnoid cysts, previous head trauma, static cerebral encephalopathy, CNS infections, CNS radiation, hydrocephalus, meningomyelocele, and neurodevelopmental disabilities may also be associated with CPP. For hypothalamic-pituitary disorders, the endocrine symptoms of precocious puberty may precede neurological symptoms. Three predictors associated with CNS lesions in girls are: (1) age younger than 6 years; (2) absence of pubic hair; and (3) estradiol concentrations greater than 110 pmol/L. The type of CNS lesion influences the presentation of GnRH-dependent precocious puberty presumably due to differences in the mechanisms inducing puberty and to the hypothalamic-pituitary deficiencies associated with the initial lesion or its treatment.

Optic gliomas are associated with neurofibromatosis type 1 (NF1), an autosomal-dominant disorder diagnosed based on clinical features that include size and number of café-au-lait spots, macrocephaly, and family history of the disorder. The mechanism of action has been presumed to be a mass effect. Pineal tumors can be associated with precocious puberty due to tumor-induced hydrocephalus or germ cell tumors. CNS radiation, dose range 18 to 50 Gray, used to treat intracranial tumors or used prophylactically for malignancies, can induce precocious gonadarche, perhaps as a result of an astroglial response with increased TGFα production. In the situation of CPP, simultaneous GH deficiency (GHD) may be masked by the accelerated growth velocity associated with increased gonadal sex steroid secretion.

Other situations associated with progressive CPP include children with virilizing disorders (such as CAH) and familial male-limited precocious puberty (testotoxicosis) in whom skeletal maturation is usually markedly advanced. In these situations, the precocious gonadarche is considered to be secondary to the virilizing disorder, but the mechanism through which the GnRH pulse generator is prematurely activated is unclear.

Disorders of Steroidogenesis

Approach to the Child With Precocious Pubertal Development

Identification of the etiology of sexual precocity begins with a thorough medical history, with questions focusing on the physical manifestations of puberty and the age and sequence of their appearance ( Figs. 17.12 and 17.13 ). Does the patient have neurological symptoms or gelastic seizures? Has the patient received radiation therapy? Was the patient exposed to exogenous hormones? Obtaining a complete family history is important because some disorders, such as familial male-limited precocious puberty, glucocorticoid resistance, and NF1, show autosomal-dominant inheritance.

Algorithm for initial approach to precocious puberty in a girl.

Beginning with a thorough medical history, this flowchart provides guidelines to the differential diagnosis of the most common causes of precocious puberty in girls. 17-OHP , 17-Hydroxyprogesterone; ACTH , adrenocorticotropin; BA , bone age; CAH , congenital adrenal hyperplasia; DHEAS , dehydroepiandrosterone sulfate; fT4 , free thyroxine; GnRH , gonadotropin-releasing hormone; LH , luteinizing hormone; NF1 , neurofibromatosis type 1; TSH , thyroid-stimulating hormone.

Algorithm for initial approach to precocious puberty in a boy.

Beginning with a thorough medical history, this flowchart provides guidelines to the differential diagnosis of the most common causes of precocious puberty in boys. 17-OHP , 17-Hydroxyprogesterone; ACTH , adrenocorticotropin; BA , bone age; CAH , congenital adrenal hyperplasia; DHEAS , dehydroepiandrosterone sulfate; FSH , follicle-stimulating hormone; GnRH , gonadotropin-releasing hormone; hCG , human chorionic gonadotropin; LH , luteinizing hormone; UFC , urinary free cortisol.

Next, a complete physical examination, including auxological measures, should be performed; particular attention should be paid to development of secondary sexual characteristics. For girls, are breast and/or pubic hair development present? For boys, is testicular volume increased? Because increased growth velocity occurs concomitantly with breast development among girls, observation of accelerated growth velocity helps to differentiate GnRH-dependent precocious gonadarche from nonprogressive gonadarche or premature thelarche. The physical examination should be directed to detect physical signs indicating onset of gonadarche or adrenarche, as well as features suggestive of specific disorders such as café-au-lait spots in MAS or NF1. Does the patient have manifestations of hypothyroidism?

Delayed Puberty

In most instances, delayed puberty is the result of delayed gonadarche and is defined as development of secondary sexual characteristics at a chronologic age greater than two SDs above the normal population. Thus for girls, lack of breast development by age 13 years or menarche by 16 years may be viewed as delayed. For boys, lack of increased testicular volume at age 14 years is considered to be delayed. Pubertal delay can be broadly subclassified as constitutional, hypogonadotropic, or hypergonadotropic.

Atypical, delayed, or absence of pubertal development during the adolescent years may also reveal disorders of sexual differentiation that escape diagnosis during infancy. Failure to complete puberty within 5 years also warrants evaluation. In general, treatment of delayed puberty involves steroid hormone replacement, which is discussed in detail at the end of the section ( Box 17.2 ; Table 17.3 ).

Box 17.2

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Gonadotropin Hormones and Their Receptors Steroid Hormones and Other Lipid Molecules Involved in Human Reproduction Structure, Function, and Evaluation of the Female Reproductive Tract Hormone-Responsive Cancers Medical Approaches to Ovarian Stimulation for Infertility Pelvic Imaging in Reproductive Endocrinology

Stay updated, free articles. Join our Telegram channel

Join